The subject disclosure generally relates to voice search services, and, in particular, to improving the performance and usability of voice search.
Voice search, also called voice-enabled search, allows the user to use a voice command to search the Internet, or a portable device. When a user is performing a voice search, the user inputs a voice query and the system interprets the voice input by the user and provides the user with search suggestions according to the interpretation. However, for various reasons, including the accent of a user, the system may incorrectly interpret the voice input of the user. If the user's utterance is incorrectly understood by the system, the user's natural response is to just click the cursor within the search box and begin typing in the intended search criteria using text based search. It may be possible for the user to reenter their voice query, but this may require extra steps and users usually may be discouraged after trying the voice input option once and not receiving desired results.
Thus, it may be desirable to provide a user with alternative voice interpretation results upon detecting that the user has not received satisfactory results based on their voice input.